arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2501.02031cs.CLcs.AI
收藏

CarbonChat:基于大语言模型的企业碳排放分析与气候知识问答系统

CarbonChat: Large Language Model-Based Corporate Carbon Emission Analysis and Climate Knowledge Q&A System

College of Future Information Technology(未来信息技术学院);Shijiazhuang University(石家庄大学)

Zhixuan Cao, Ming Han, Jingtao Wang, Meng Jia

AI总结:

本文提出 CarbonChat 系统,通过多样化索引、自提示 RAG、Text2SQL、14维碳分析及幻觉检测,实现企业碳排放分析与气候知识问答。

摘要:

随着全球气候变化影响加剧,企业碳排放已成为全球关注焦点。针对大语言模型中气候变化知识更新滞后、传统增强生成架构在复杂问题上缺乏专业性与准确性,以及可持续发展报告分析成本高、耗时长等问题,本文提出 CarbonChat:一个基于大语言模型的企业碳排放分析与气候知识问答系统,旨在实现精准的碳排放分析与政策理解。首先,提出多样化索引模块构建方法,用于处理基于规则文档和长文本文档的切分,以及结构化数据提取,从而优化关键信息解析。其次,设计增强型自提示检索增强生成架构,融合意图识别、结构化推理链、混合检索和 Text2SQL,提升语义理解与查询转换效率。接着,基于温室气体核算框架建立14个碳排放分析维度,实现报告摘要、相关性评估和定制化回答。最后,通过多层分块机制、时间戳和幻觉检测功能,确保分析结果的准确性与可验证性,降低幻觉率并提高回答精度。

英文摘要:

As the impact of global climate change intensifies, corporate carbon emissions have become a focal point of global attention. In response to issues such as the lag in climate change knowledge updates within large language models, the lack of specialization and accuracy in traditional augmented generation architectures for complex problems, and the high cost and time consumption of sustainability report analysis, this paper proposes CarbonChat: Large Language Model-based corporate carbon emission analysis and climate knowledge Q&A system, aimed at achieving precise carbon emission analysis and policy understanding.First, a diversified index module construction method is proposed to handle the segmentation of rule-based and long-text documents, as well as the extraction of structured data, thereby optimizing the parsing of key information.Second, an enhanced self-prompt retrieval-augmented generation architecture is designed, integrating intent recognition, structured reasoning chains, hybrid retrieval, and Text2SQL, improving the efficiency of semantic understanding and query conversion.Next, based on the greenhouse gas accounting framework, 14 dimensions are established for carbon emission analysis, enabling report summarization, relevance evaluation, and customized responses.Finally, through a multi-layer chunking mechanism, timestamps, and hallucination detection features, the accuracy and verifiability of the analysis results are ensured, reducing hallucination rates and enhancing the precision of the responses.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑